CN106663960B

CN106663960B - Battery system

Info

Publication number: CN106663960B
Application number: CN201580032416.4A
Authority: CN
Inventors: 井上健士; 山田惠造; 松本哲也
Original assignee: Showa Denko KK
Current assignee: Resonac Holdings Corp
Priority date: 2014-06-30
Filing date: 2015-06-15
Publication date: 2021-01-15
Anticipated expiration: 2035-06-15
Also published as: CN106663960A; KR101925196B1; EP3163710A1; JP6245094B2; EP3163710B1; US10461545B2; US20170141589A1; EP3163710A4; WO2016002485A1; KR20170008807A; JP2016012984A

Abstract

In a battery system in which a plurality of types of batteries are connected via a Switch (SW), when the Switch (SW) is switched in accordance with a voltage (charging rate), a charging charge during charging may be further increased in accordance with a direct-current resistance of the battery. The invention provides a Switch (SW) switching control for further increasing the charge. A first feature of the present invention is that, in a battery system in which a first battery and a second battery are connected in parallel via a Switch (SW), a means for measuring the resistance and OCV of each battery to estimate a charging current is required, and the charging charge can be further increased by switching to a combination of Switches (SW) that increases the charging charge.

Description

Battery system

Technical Field

The present invention relates to a battery system for supplying electric power to an electric load, and to an electric storage system including 2 secondary batteries.

Background

As an example of an energy monitoring system, in recent years, in addition to an idle stop function, a micro hybrid vehicle (hereinafter, referred to as a micro HEV) has been developed in which regenerative energy during deceleration is converted into electric energy by a generator (ac generator) to charge a battery, and the battery is used as a power source for auxiliary devices such as headlights and heaters as an electrical load. Here, in recent micro-HEVs, a lead battery and another battery (hereinafter referred to as a sub-battery) are sometimes used as a secondary battery. The aim is to recover more renewable energy. Since the Open Circuit Voltage (OCV) of the sub-battery used in the 3 types of vehicles is substantially the same as that of the lead battery, even when 2 batteries, that is, the lead battery and the sub-battery, are connected in parallel, the exchange of current (hereinafter, referred to as "cross current") between the batteries can be prevented.

However, when a battery having an OCV different from that of the lead battery is used as the sub-battery, a loss occurs due to the cross current, and the regenerative energy cannot be sufficiently recovered. In particular, when an electric storage device (not a battery but an electric storage device, but described herein as a battery) capable of increasing a charging current and having excellent temperature resistance and life is used in the sub-battery, the cross current becomes significant.

In order to prevent the lateral current, a DCDC converter is interposed between the lead battery and the sub-battery, but the cost increases. Therefore, in order to reduce costs (not limited to micro-HEVs), the following method is considered: as a hardware structure, by the method disclosed in patent document 1 (japanese patent laid-open No. 2010-115050), switches SW are inserted in series and connected in parallel to the lead battery and the sub-battery, respectively, thereby preventing the cross current.

As a method for switching the switch SW, patent document 1 discloses a method for switching the switch SW so that the charging rates of the main battery and the sub battery are equalized. Further, patent document 2 discloses a switching method based on a voltage change.

Patent document 1: japanese laid-open patent application No. 2010-115050

Patent document 2: japanese patent No. 3716776

Disclosure of Invention

However, as in patent document 1, when the charging rates of a lead battery (hereinafter, referred to as a first battery) and a sub-battery (hereinafter, referred to as a second battery) are determined based on the values of voltage and current, and a battery to be charged is determined based on the result of the determination, it is possible that the amount of charge is determined to be sufficient even if the amount of charge is insufficient, depending on the resistance of the battery, and in this case, the battery has a room for being able to absorb the charge. In particular, when charging is switched between a plurality of batteries having different properties such as capacity and resistance (for example, a lead storage battery and a lithium ion secondary battery) as in a micro-HEV, there arises a problem that only the lithium ion secondary battery is charged. That is, even if the voltage of the lithium ion secondary battery rises, the resistance is lower than that of the lead storage battery, and therefore, the battery is not switched to the lead storage battery, and there is a room for increasing the total amount of charge to the first battery and the second battery.

The purpose of the present invention is to provide a battery system that can increase the total charge amount even in cases such as a first battery and a second battery that are different in properties.

The present invention includes the following.

A battery system in which a first battery and a second battery are connected in parallel via a switch SW, the battery system comprising means for estimating a charging current of the first battery based on at least an internal resistance of the first battery and means for estimating a charging current of the second battery based on at least an internal resistance of the second battery, wherein the switch SW is switched so that a sum of charging charges to the first battery and the second battery becomes larger based on the charging current of the first battery and the charging current of the second battery.

In the battery system, the switch SW is a first switch SW and a second switch SW, the first switch SW and the second switch SW are connected in parallel, the first battery is connected to a load via the first switch SW, and the second battery is connected to the load via the second switch SW.

In the battery system, the second battery is discharged first when discharging, and the first battery is switched to discharge when the second battery has reached a predetermined voltage or a predetermined charge rate, or the first battery is discharged first and the second battery is switched to discharge when the first battery has reached a predetermined voltage or a predetermined charge rate.

In the battery system, the switch SW is switched once during regenerative charging, and when the regenerative time is T, the charging time of the second battery is τ, and the first charging time is T- τ, where T > τ, the second battery is first charged, and the switch SW is switched to the first battery at the timing τ at which the amount of charge of the first battery and the second battery becomes maximum.

In the battery system, τ is a time series of current estimated for the first battery (i1(t)) and a time series of the second battery (i2(t)), and the second battery is initially charged at time t from the start of charging when i1(t) is i2 (charging end time-t), and the switch SW is switched so as to select the first battery after t has elapsed from the start of charging.

In the battery system, when the regenerative charging time T is unknown, the time T at which the current timing of the first battery becomes equal to the current convergence value of the second battery is determined as τ.

In the battery system, the switch SW is switched a plurality of times during regenerative charging, and the estimated charging current of the first battery and the estimated current of the second battery are compared to select the battery having a large estimated current, thereby switching the charging of the switch SW.

In the battery system, the switch SW is periodically switched and charged until one of the first battery and the second battery is charged by a constant voltage.

In the battery system, when the first battery becomes a constant-current charge state, the time ratio of the switch SW of the first battery is first set to 1, and after the first battery becomes a constant-current charge end state, the time ratio of the switch SW of the first battery is set to (voltage at constant-voltage charge of the ac power generator-open-circuit voltage of the first battery)/(current at constant-current charge of the ac power generator × polarization resistance).

In the battery system, the load is an alternating current generator, and when the first battery becomes a constant current charge, the time ratio of the switch SW of the first battery is controlled so that a current at the time of the constant current charge of the alternating current generator is equal to a polarization voltage of the second battery/(polarization resistance of the second battery (1+ polarization capacitance of the second battery/capacitance of the second battery)) + a polarization voltage vp (t)) of the first battery/polarization resistance of the first battery.

In the battery system, the switch SW is switched a plurality of times during regenerative charging, and when the switch SW is switched a plurality of times during charging and simultaneous connection is permitted, the charging currents of the first battery and the second battery are estimated in 3 cases when the first battery alone, the second battery alone are charged and both the first battery and the second battery are connected, and when the first battery or the second battery becomes discharged, the switch SW is switched so as to connect the first battery and the second battery having a large charging current, otherwise, when the battery alone becomes charged at a constant voltage, both the first battery and the second battery are connected, otherwise, the on time of the switch SW alone of the first battery, the on time of the switch SW alone of the second battery, and the ratio of the switch SW time for connecting both the batteries are controlled, so as to become a constant current charge.

In the battery system, a ratio of an on time of the switch SW for the first battery alone, an on time of the switch SW for the second battery alone, and a time of the switch SW for connecting both batteries is set to a predetermined value.

In the battery system, as the ratios of the on time of the switch SW for the first battery alone, the on time of the switch SW for the second battery alone, and the switch SW time for connecting both batteries, the ratio of the switch SW time for connecting both batteries is set to 0, the time ratio of the switch SW for the first battery is initially set to 1, and the ratio of the switch SW time for the first battery after the first battery has reached the constant-current charge complete state is set to (voltage at constant-voltage charging of the ac generator-open-circuit voltage of the first battery)/(current at constant-current charging of the ac generator x polarization resistance).

In the battery system, the ratio of the switch SW time for connecting both batteries is set to 0 as the ratio of the switch SW on time for the first battery alone, the switch SW on time for the second battery alone, and the switch SW time for connecting both batteries, and the time ratio of the switch SW for the first battery is controlled so that the current during constant current charging of the ac generator becomes the polarization voltage of the second battery/(polarization resistance of the second battery (1+ polarization capacitance of the second battery/capacitance of the second battery (F equivalent)) + the polarization voltage of the first battery/polarization resistance of the first battery).

The battery system is characterized by comprising a unit for measuring the voltage and current of the first battery, the voltage and current of the second battery, and the voltage of the alternating current generator and auxiliary machines, and measuring the direct current resistance of the first battery, the resistance of the switch SW, the polarization capacitance, the polarization resistance, the polarization voltage, the open circuit voltage, and the capacitance of the second battery.

In the battery system, the dc resistance of the battery and the resistance of the switch SW are determined based on changes in current and voltage before and after the switch SW is changed from on to off or the switch SW is changed from off to on, and the open-circuit voltage of the battery is determined as the voltage-dc resistance × current of the battery.

In the battery system, the parameters are estimated on-line from the current time series and the voltage time series of the battery so that the polarization resistance, the polarization capacitance, and the capacitance of the second battery conform to a circuit equation assumed in advance.

In the battery system, it is estimated that the polarization voltage is one polarization voltage x (1-measurement time step/(polarization resistance × polarization capacitance)) + measurement current × measurement time step/polarization capacitance before the measurement time.

In the battery system, the open circuit voltage of the battery in the steady state is set as the measured open circuit voltage-polarization voltage.

In a battery system, a relationship between an open circuit voltage of a battery and a charging rate of the battery, which are in a stable state of the battery, is held in advance, an initial charging rate of the battery is obtained from a voltage at the time of system start, the charging rate is updated by adding a value of current integration/a capacity (Ah) of the battery, and the open circuit voltage in the stable state is obtained from the charging rate.

In the battery system, the polarization voltage of the battery is set to a measured voltage of the battery, i.e., a direct current resistance x a measurement current, and an open circuit voltage of the battery in a steady state.

In the battery system, a current that can be charged in the battery system is transmitted to the higher-level controller via the communication line.

In the battery system, a charge start signal, a current during constant current charging, and a voltage during constant voltage charging are transmitted from a higher-level controller to the battery system via a communication line.

In the battery system, the time from the start to the end of charging is transmitted from the higher-level controller to the battery system together with the start of charging via the communication line.

According to the present invention, it is possible to provide a battery system capable of increasing the total charge amount even in the case of the first battery and the second battery having different properties.

Drawings

Fig. 1 is a schematic configuration diagram showing a micro HEV equipped with a battery system to which the present invention is applied.

Fig. 2 is a schematic configuration diagram showing a battery system according to the present invention.

Fig. 3 is a diagram showing the overall process of a battery system to which the present invention is applied.

Fig. 4 is a diagram illustrating processing of the battery system at the time of non-regeneration.

Fig. 5 is a diagram showing a process in startup.

Fig. 6 is a diagram showing an equivalent circuit in the case where batteries are connected in parallel.

Fig. 7 is a diagram showing an example of a table of open circuit voltages of batteries.

Fig. 8 is a diagram showing an equivalent circuit of the battery.

Fig. 9 is a diagram showing an example of a table of polarization resistance/capacitance of a battery.

Fig. 10 is a diagram showing an example of a table of dc resistance of a battery.

Fig. 11 is a diagram showing a table example of the switch SW.

Fig. 12 is a diagram in which an instantaneous stop counter capacitor is added to a power supply system.

Fig. 13 is a diagram showing the logic gate processing of the switch SW.

Fig. 14 is a diagram of a corresponding process when the battery voltage is lowered after long-term parking.

Fig. 15 is a diagram showing an example of switching control during regenerative charging.

Fig. 16 is a diagram showing an equivalent circuit of a large-capacity battery.

Fig. 17 is a diagram showing an equivalent circuit of a small-capacity battery.

Fig. 18 is a diagram showing an equivalent circuit of the electric storage device.

Fig. 19 is a diagram showing an image of the geometry for calculating the battery switching time.

Fig. 20 is a diagram showing another example of the switching control at the time of regenerative charging.

Fig. 21 is a diagram showing an example of the switch SW selection process during constant current charging.

Fig. 22 is a diagram showing still another example of the switching control at the time of regenerative charging.

Fig. 23 is a diagram showing examples of factors of the battery.

Fig. 24 is a graph showing comparison of effects of regenerating the charge.

Fig. 25 is a diagram showing embodiment 4 of switching control at the time of regenerative charging.

(symbol description)

11: an engine; 12: generators (alternating current generators); 14: an auxiliary machine load; 15: an ECU (upper controller); 16: a communication line; 17: a micro-HEV; 200: a controller; 201: a first battery; 202: an ammeter of the first battery; 203: a voltage sensing line of the first battery; 204: a current estimation unit for the first battery; 205: a switch SW of the first battery; 206: a second battery; 207: an ammeter of the second battery; 208: a voltage sensing line of the second battery; 209: a current estimation unit for the second battery; 210: a switch SW of the second battery; 211: a comparator; 212: an SW (switch) control unit; 213: an alternator (alter)/auxiliary voltage sense line; 31: setting a switch SW position in the parking process; 32: a controller dormancy step; 33: ignition-on determination; 34: a controller awakening step; 35: a regeneration start judgment; 36: a regenerative charging control step; 37: a non-regeneration control step; 38: ignition off determination; 41: a determination in startup (cranking); 42: starting a middle processing step; 43: starting to judge; 44: starting a starting instruction; 45: a charging disconnection instruction step of the alternating current generator; 46: a determination of whether the second battery is empty; 47: a first battery discharge instruction step; 48: a second battery discharge instruction step; 501: simultaneously judging connection; 502: judging whether the voltage of the auxiliary engine of the alternator is below a threshold value; 503: a cross flow generation decision; 504: cross flow generation determination during simultaneous connection; 505: connecting instructions simultaneously; 506: comparing and judging the voltage when the batteries are connected independently; 507: a connection instruction step of only the first battery; 508: a connection instruction step of only the second battery; 509: the second battery can be used alone for judgment; 510: a connection instruction step of only the first battery; 511: a connection instruction step of only the second battery; 81: a polarization capacitance; 82: a polarization resistance; 121: backup capacitor 1301: a first battery turn-on signal; 1302 a: a second battery turn-on signal; 1302 b: an OR logic gate for judging disconnection of the two batteries; 1303: NOT logic gate for judging disconnection of the two batteries; 1304: a two-battery disconnection determination signal; 1305: the first battery is connected with the extended OR logic gate; 1306: a delay circuit; 1307: a delay circuit; 1308: the first battery is connected with the extended OR logic gate; 1309: the first battery guarantee uses an OR logic gate; 1310: a pull-up resistor; 1311: a first battery gating signal; 1312: a second battery strobe signal; 141: a first battery voltage threshold insufficiency determination; 142: a second battery voltage threshold insufficiency determination; 143: a first battery only on command step; 144: a step of sending a start instruction; 145: determining that the first battery OCV threshold value is higher than or equal to a first battery OCV threshold value; 146: determining whether the voltage of the second battery is higher than a threshold value; 147: a second battery only on command step; 148: an engine stop instruction sending step; 151: a second battery selection instruction step; 152: receiving a first battery presumed current time series; 153: receiving a current-presumed time series by the second battery; 154: calculating switching time; 155: judging the regeneration elapsed time switching time; 156: judging the end of regeneration; 157: a second battery selection instruction; 158: judging the end of regeneration; 171: a small-capacity battery OCV; 191: a second battery current time series; 192: a first battery current time series; 193: switching time; 2001: a first battery estimated current receiving step; 2002: a second battery estimated current receiving step; 2003: a determination whether CC charging is possible even when the first battery/second battery is used alone; 2004; comparing and judging the first battery presumed current and the second battery presumed current; 2005: a first battery selection instruction step; 2006: a second battery step; 2007: judging the end of regeneration; 2008: selecting and processing a switch SW during CC charging; 2010: judging the estimated current CC of the first battery is insufficient; 2102: a first battery selection instruction; 2013: a second battery selection instruction; 2201: a first battery estimated current receiving step; 2202: a second battery estimated current receiving step; 2203: receiving the battery speculated current when the connection is carried out simultaneously; 2204: a determination whether CC charging is possible even when the first battery/second battery is used alone; 2205: selecting and processing a switch SW during CC charging; 2206: cross flow generation determination during simultaneous connection; 2207: judging the end of regeneration; 2208: a first battery inferred current and a second battery inferred current comparison; 2209: a first battery selection instruction step; 2210: simultaneously connecting two batteries; 2211: a second battery selection instruction step; 251: a first battery estimated current receiving step; 252: a second battery estimated current receiving step; 253: receiving the battery speculated current when the connection is carried out simultaneously; 254: cross-flow determination during simultaneous connection; 255: connecting the two batteries; 256: a second battery undervoltage determination; 257: a second battery selection instruction; 258: a first battery selection instruction; 259: and (4) judging the end of regeneration.

Detailed Description

The embodiments of the present invention will be described in detail below with reference to the drawings, but the present invention is not limited to the embodiments below, and various modifications and application examples are included in the technical concept of the present invention. For example, the embodiments described below can be applied to HEVs and XEMS (HEMS and BEMS), and also to power storage systems mounted on electric vehicles and railways, by changing the battery voltage.

Fig. 1 is a schematic diagram of the structure of a micro-HEV.

In fig. 1, a micro-HEV 17 on which a battery system 10 (a pack including 2 types of secondary batteries, for example, a lead battery and another battery including an electric storage device, which will be described later) is mounted includes an engine 11, an auxiliary load 14 as an electrical load, such as a generator 12 (an ac generator) mechanically connected to the engine 11, a lamp, an air conditioner fan, and a starter, an ECU15 as an upper controller, and a communication line 16.

Here, during idling, electric power for auxiliary loads 14 of micro-HEV 17 is supplied from battery system 10. Further, the ac generator 12 is rotated and operated by a rotational force (deceleration energy) from the tires generated by the coasting operation of the vehicle when the vehicle decelerates, and the electric energy generated by the ac generator 12 is supplied as electric power to the auxiliary load 14, and the secondary battery in the battery system 10 is charged. Here, the voltage of ac generator 12 is set to a rated voltage (for example, 14V) of auxiliary load 14. Furthermore, ac power generators are typically constant current power sources. However, when the voltage reaches a predetermined value, the voltage is controlled to a constant voltage (14V in the case of a normal vehicle) by control. That is, the ac generator is regarded as a CCCV (Constant Current Constant Voltage) charger. The ECU controls the mechanical brake and the on/off of the ac generator during regeneration, and sends a charging current to the power supply system.

The battery system 10 includes a first battery (generally, a lead battery) 201, an ammeter 202 that monitors a current of the first battery, a voltage sensing line 203 that monitors a voltage of the first battery, a current estimating unit 204 that estimates a current of the first battery, a switch SW205 connected in series with the first battery, a second battery 206, an ammeter 207 that monitors a current of the second battery, a voltage sensing line 208 that monitors a voltage of the second battery, a current estimating unit 209 that estimates a current of the second battery, a switch SW210 connected in series with the second battery, a comparator 211 that compares the currents, a switch control unit 212 that generates a signal for controlling the switch SW, and a voltage sensing line 213 of an alternating-current generator/auxiliary machine.

The controller 200 includes a switch control unit 212, a comparator 211, and

current estimation units

204 and 209. The controller 200 receives the current and voltage of each battery and information from the upper controller 16 via the signal line 18, and controls the on/off states of the switches SW205 and SW 210. By switching the switch SW, the current from the generator 12 can be switched to flow into the first battery or flow into the second battery.

Here, a component of the signal directly controlling the switch SW is a switch control unit 212 in the controller 200. The power source of the controller 200 may also use the first battery. The switch SW205 and the switch SW210 are inhibited from being turned off, thereby preventing the auxiliary power supply from being lost. Further, the switches SW205 and SW210 may be power MOS-FETs, IGBTs, or mechanical relays.

With the above configuration, first battery 201 and second battery 206 are discharged in a certain state so as to supply electric power to auxiliary load 14, and are charged with electric energy generated by generator 13 at the time of deceleration of micro-HEV 17.

As the first battery 201 and the second battery 206, a lead battery, a nickel-metal hydride battery, a nickel-zinc battery, a lithium ion battery, an electric double layer battery, a lithium ion battery, or the like can be used.

Generally, a lead battery is used as the first battery 201. The reason for this is that a battery having a large capacity (Ah) is required to secure electric power for the protection device and the like during long-term parking.

In addition, a battery designed to be able to charge a large amount of normal current, a lithium ion battery, a nickel zinc battery, or a nickel hydrogen battery may be used as the second battery 206. Since the rated voltage of a lithium ion electric storage device or a lithium ion battery is generally in the range of 3V to 4.2V, 4 devices are used in series in order to converge to the range of 8V (reference of sound pressure) to 14V, which is a reference of the vehicle voltage range. In the case of a nickel-metal hydride battery, 10 pieces were used. In the case of nickel-zinc batteries, 8 to 10 blocks were used, and in the case of accumulators, 7 blocks were used.

As described above, generally, the 1 st battery 201 uses a capacity type battery or an electric storage element such as a lead battery with a high capacity (Ah) and the 2 nd battery 206 uses a power type battery capable of generating electric power such as a lithium ion battery or an electric storage device, but the present invention is not limited thereto, and the 1 st battery 201 and the 2 nd battery 206 may use the same type of battery.

As the ammeter 202 of the first battery and the ammeter 207 of the second battery, a hall element and a shunt ammeter can be used.

In the present invention, in a battery system in which the first battery 201 and the second battery 206 are connected in parallel via the switch SW, for example, in the above-described battery system 10, a means for measuring the resistance and OCV of each battery to estimate the charging current is required, and the combination of on/off switching can be switched so that the total charge of each battery becomes large.

The switch SW is switched not based on the result of the ammeter provided in each battery, but based on information obtained by means for estimating the charging current by measuring the resistance and OCV of each battery, the switch SW is switched for the first battery 201 and the second battery 206, so that it is possible to provide a battery system capable of increasing the total amount of charge even in the case of the first battery and the second battery having different properties such as resistance and capacity.

Here, the following 4 embodiments are described when determining such a combination of switches SW that the charge becomes large. Further, the user can select the following 4 embodiments.

In the first embodiment, at the time of regeneration, first, either one of the first battery 201 and the second battery 206 is selected to be charged, then the switch SW is switched to charge the other battery, and the switching of the switch is performed only once during a period from the start to the end of regeneration. As the timing of switching, for example, a current time series in the case where only one cell is connected is estimated, and when the current time series of the first cell is represented by I (T) and the current time series of the second cell is represented by I (T), the time series I (T- τ) and I (τ) are compared to find the solution τ of the nonlinear equation so that they become equal to each other, and when τ has elapsed from the regeneration, the second cell is switched to the first cell (T is the regeneration time).

In the second embodiment, the switch is switched a plurality of times during a period from the start to the end of regeneration. The estimated currents of the first battery and the second battery are obtained at a constant interval, and the switch is switched to the larger one of the estimated currents of the first battery and the second battery each time. Here, when the battery is in the constant current charging mode, the current is the same regardless of which battery is connected, and therefore, the following may be used. When the battery is opened, the battery whose voltage is rapidly decreased (hereinafter, the battery whose polarization is rapidly released) is first subjected to CC charging. Then, the first battery and the second battery are alternately switched by the switch SW so that the OCV of the battery whose polarization release is fast becomes constant. This makes it possible to extend the CC charging time and increase the charge by utilizing the recovery due to the decrease in the battery voltage caused by the polarization release.

In the third embodiment, the switch is switched a plurality of times during the period from the start to the end of regeneration, and both batteries may be connected so as to be charged simultaneously. After the CC charging of the second embodiment is completed, the switches of the 2 batteries are turned on. In this case, the cross current is always monitored, and when the cross current is generated, the cross current can be prevented by combining control such as turning on the switch SW of only 1 cell whose charging current is larger immediately.

In the fourth embodiment, the first switch and the second switch may be turned on to connect both the first battery and the second battery, but the number of times of switching of the switch SW may be reduced. After CC charging according to the third embodiment, first, one battery is charged to a voltage at which no cross current is generated even when the batteries are connected in parallel. Then, if the cross current does not occur even in parallel, the parallel connection is established.

In the case where the first battery and the second battery are not simultaneously connected as in

embodiments

1 and 2, one switch may be provided at one intersection of the first battery and the second battery connected in parallel, but a plurality of switches may be provided. In contrast, in examples 3 and 4 in which simultaneous connection is permitted, the switches are a first switch and a second switch, the first switch SW and the second switch SW are connected in parallel like the first battery and the second battery, the first battery is connected to the load (the auxiliary load 14, the ac power generator 12, and the like) via the first switch SW, and the second battery is connected to the load via the second switch SW.

In the present invention, both batteries are not turned on to prevent a cross current at the time of discharging, and during the next regeneration, the regenerative charge is increased to decrease the OCV of the second battery, and at the time of discharging, the second battery is discharged first, and when the second battery reaches a predetermined voltage or charge rate, the second battery is switched to the first battery to discharge, or the first battery is discharged first, and when the first battery reaches a predetermined voltage or charge rate, the second battery is switched to discharge, but when the engine is started (started), a current of 300A is necessary, and therefore, there is a possibility that power is not generated only in the first battery and the engine cannot be started. In this case, the first battery and the second battery may be connected in parallel to supplement the power shortage, or the second battery may be switched to be connected to the first battery.

Hereinafter, examples 1 to 4 will be described in detail including the overall control.

Example 1

Fig. 3 shows an outline of the overall process of control from ignition on to off and parking after factory shipment or after first battery replacement.

First, it is considered that the switch SW205 is turned on and the switch SW210 is turned off during parking in step 31, the controller 200 is put to sleep in step 32, and the first battery is charged with electric power of the protection device during parking and shifts to the low power consumption mode.

This process continues until the ignition switch becomes on. If the ignition is turned on at decision 33, the process proceeds to step 34 and the controller is awakened.

Next, in step 35, it is determined whether or not to start regeneration. This determination is determined by a signal from the ECU 16. After the regeneration is started, the process proceeds to the regenerative charging control at step 36. Details thereof are described in other embodiments. When the processing ends at step 36, the processing proceeds to non-regeneration control step 37. Then, in the determination of step 39, if the ignition-on is continued, the process is shifted to step 35, and if the ignition-off is turned, the process is shifted to step 31, which is the process during parking. Further, information from the ECU16 is obtained with respect to a signal of an ignition switch.

The processing of step 35 to step 39 may be executed or determined periodically in accordance with an event of a control cycle (for example, 10ms, 0.1 s). In addition, with respect to the process of fig. 3, the process continues until the vehicle is scrapped or the first battery is replaced.

In fig. 4, the non-regeneration control 37 in fig. 3 is explained. The non-regenerative control 37 is divided into two modes, a discharge mode and a forced charge mode. The discharge mode is divided into three processes of a case where the charge charged in the second battery after the regenerative charging is large, a case where the second battery becomes empty and only the first battery is used, and a start. The forced charge mode is a case where both batteries are empty and the first battery is forcibly charged. In the example of fig. 4, this process is explained. Further, in the course of regeneration or in the case where regeneration is started, the processing of fig. 4 is not executed.

First, in step 41, it is determined whether the engine is in a start. If it is the startup, the process is transferred to a process-on-startup 42, and if it is not, the process is transferred to a step 43. Information is obtained from the ECU15 as to whether or not it is in the start.

In step 42, the switch SW processing in startup is executed. This process will be described later. After the end of step 42, the process of fig. 4 is ended.

In step 43, it is determined whether or not the start-up is started. If start-up is initiated, the process is transferred to step 44. If the start is not started, a charging off command of the ac generator is sent to the ECU15 in step 45 (for the purpose of preventing deterioration of fuel consumption due to the output UP of the excessive ac generator from the viewpoint of fuel consumption saving). Here, the start is based on the cause of the vehicle (for example, when the room temperature rises during idling due to the stop of the air conditioner compressor during idling), and this means that the first battery becomes empty in the battery system 10 and the first battery needs to be forcibly charged. Information is obtained from the ECU15 regarding the start of startup due to vehicle-side reasons. The determination as to whether or not the first battery is empty may be performed when the charging rate of the first battery is equal to or less than a predetermined charging rate. When a lead battery is used as the first battery, the predetermined charging rate may be, for example, 80% or 90%. As another determination method as to whether or not the first battery is empty, the voltage of the first battery may be set to a predetermined voltage or lower. The predetermined voltage may be 12.4V or 12.6V. In addition, regarding the charge interruption of the ac generator, the ac generator may be stopped, or the generated voltage of the ac generator may be adjusted to the OCV of the battery (in this case, the generated voltage adjusting function of the ac generator is required). As the generated voltage adjustment of the ac generator, the current (sum of 202 and 206) measured by the ammeter may be transmitted to the ECU15, and the ECU may perform voltage control using feedback. After the end of step 45, the process is transferred to step 46.

In step 44, the switch SW205 of the first battery is turned on, and the switch SW210 of the second battery is turned off to prepare for activation. In addition, the current state of the switch SW may be continued. After the processing of step 44, the processing of fig. 4 is ended.

In step 46, it is determined whether the second battery is empty. As a determination method, the voltage of the second battery may be equal to or lower than a predetermined voltage. The preset voltage may be a rated voltage of the second battery × the number of the second batteries connected in series, or may be a voltage causing an acoustic sound jump, for example, 8V. In the case where the second battery is empty, the process is transferred to step 47. In the case where the second battery is not empty, the process is transferred to step 48.

In step 47, to discharge the first battery, the switch SW205 is turned on and the switch SW210 is turned off. After step 47 ends, the flow of fig. 4 ends.

In step 48, the switch SW205 is turned off and the switch SW210 is turned on in order to discharge only the second battery. This is a process for emptying the second battery as much as possible at the time of discharge in order to absorb the charge current as much as possible in the next regeneration. After the end of step 48, the flow of fig. 4 is ended.

The above flow of fig. 4 is an example in which the second battery is discharged first, but the first battery may be discharged first.

Fig. 5 is a diagram illustrating the in-startup processing 42 in fig. 4. In fig. 5, the on/off state of the switch of the previous time is set to the initial value of the on/off state of the switch of fig. 5.

First, in step 501, it is determined whether or not both of the switches SW205 and SW210 are on, and if yes, the process proceeds to step 503, and if no (only one of the switches SW is on), the process proceeds to step 504.

In step 502, it is determined whether or not the alternator/auxiliary machinery voltage (measured by voltage sensing line 213 in fig. 2) is equal to or less than a predetermined threshold value. If the threshold value is not larger than the threshold value, the process proceeds to step 504, and the process of fig. 5 is terminated if not. Here, the threshold may be, for example, 8V at which a sound skipping occurs. As for the alternator/auxiliary machinery voltage, a value measured by the voltage sensing line 213 in fig. 2 is used.

In step 504, the estimated current of each battery when both switches are assumed to be on is obtained, and it is determined which current is charging (cross current). In the case where a cross flow is generated, the process is transferred to step 506 (a process for preventing a cross flow). In the case where no cross flow is generated, the process is transferred to step 505. Here, the current estimation unit 204 in fig. 2 estimates the current estimated as the determination material for the occurrence of the cross current for the first battery, and the current estimation unit 209 in fig. 2 estimates the current estimated for the second battery. The cross flow determination method will be described below.

First, the OCV (Open Circuit Voltage) of the first battery is set to V1, and the OCV of the second battery is set to V2. Then, the dc resistance of the first battery is R1, the dc resistance of the second battery is R2, the resistance of the switch SW205 is R1, and the dc resistance of the switch SW210 is R2. The method of determining this value is described later. The current required for startup is Ia. Ia may be a value sent from ECU15 or a current currently flowing in the battery (may be the sum of ammeters 202 and 206). Further, if the required power Pa is supplied instead of the current, the quadratic equation is solved by the equation of current × voltage, and the power Pa is replaced by the current × voltage

Ia is converted from Pa (in this case, Pa may be a value received from ECU15 or the current × alternator/auxiliary machine voltage, and V × V/4r ≧ Pa as a startable condition). V is the OCV of the battery and R is the sum of the dc resistance and the resistance of the switch SW. In the case where 2 cells were connected in parallel, Ia was calculated from the circuit synthesis, assuming that V ═ ((R2+ R2) × V1+ (R1+ R1) × V2)/(R1+ R2+ R1+ R2), and R ═ R1+ R1) × (R2+ R2)/(R1+ R2+ R1+ R2. V1 is the OCV of the first cell, V2 is the OCV of the second cell, R1 is the dc resistance of the first cell, R2 is the dc resistance of the second cell, R1 is the on resistance of switch SW205, R2 is the on resistance of switch SW 210.

Next, a method of estimating the charging current of the first battery 201 and the second battery 206 will be described. Since the switches are switched for the first battery 201 and the second battery 206 based on the information obtained by the means in which the resistances of the batteries are taken into consideration, a battery system capable of increasing the total charge amount even in the case of the first battery and the second battery having different properties such as the resistances can be provided.

An equivalent circuit in the case of connecting 2 batteries in parallel is represented as fig. 6. This is because the OCV that can be regarded as a battery hardly changes in the control period (e.g., 10ms) of fig. 3. From the circuit equations, the equations for the current and voltage in FIG. 3 are derived. Here, the current of the first battery is formula 1, and the current of the second battery is formula 2. In the

formulas

1 and 2, the discharge direction is expressed as +.

The current of the first cell { - (V2-V1) + (R2+ R2) Ia }/(R1+ R1+ R2+ R2) … (formula 1)

Current of the second cell { (V2-V1) + (R1+ R1) Ia }/(R1+ R1+ R2+ R2) … (formula 2)

The current estimating unit 204 can estimate the current according to equation 1, and the current estimating unit 205 can estimate the current according to equation 2. Then, if the first current is positive and the second current is positive, it is determined that no cross current is generated (conversely, if the current of the first cell × the current of the second cell <0, it is determined that cross current is generated). This determination is equivalent to the comparator 211 in fig. 2. As a simpler cross flow determination method, if expression 3 is satisfied, it may be determined that cross flow is not generated.

- (R1+ R1) or less (V2-V1)/Ia or less (R2+ R2) … (formula 3)

Next, in step 505, switch SW205 (switch SW for the first battery) is turned on, switch SW210 (switch SW for the second battery) is turned on, and both batteries are connected in parallel, whereby the processing in fig. 5 is ended.

In step 506, it is determined whether or not the first battery voltage is large based on the estimated battery voltage when the individual batteries are assumed to be connected, and if the estimated voltage of the first battery is large, the process proceeds to step 507, otherwise the process proceeds to step 508. Here, as the estimated voltage when the batteries are connected individually, the estimated voltage of the first battery is formula 4, and the estimated voltage of the second battery is formula 5. Similarly, the methods for estimating R1, R1, R2, R2, V1 and V2 will be described later.

Estimated voltage of the first cell, V1- (R1+ R1) Ia … (formula 4)

Estimated voltage of the second cell, V2- (R2+ R2) Ia … (formula 5)

Here, regarding the switch SW, if only the switch SW205 of the first battery is turned on, the value read out from the voltage sensing line 203 may be set as the estimated voltage of the first battery, and regarding the switch SW, if only the switch SW210 of the second battery is turned on, the value read out from the voltage sensing line current meter 208 may be set as the estimated voltage of the second battery.

Next, in step 507, since the voltage drop is small when the first battery is used, the switch SW205 is turned on and the switch SW210 is turned off, and the process of fig. 5 is ended.

In step 508, since the voltage drop is small when the second battery is used, the switch SW205 is turned off, the switch SW210 is turned on, and the processing of fig. 5 is ended.

Since both switches SW are currently turned on in step 503, it is determined whether or not a cross current is generated based on the values of the

ammeters

202 and 206, and if a cross current is generated, the process proceeds to step 509, and if a cross current is not generated, the state of the switch SW is maintained and the process of fig. 5 is ended.

In step 509, since the cross current is generated, it is necessary to switch to the individual battery connection, and it is determined whether or not the second battery should be used. If the second battery can be used, the process is transferred to step 511, otherwise the process is transferred to step 510. Here, the voltage of equation 5 is set to a predetermined threshold value (for example, a voltage of 8V with no audible bounce sound may be used) for the determination of whether the second battery can be used.

In step 510, since only the first battery can be used, the switch SW205 is turned on, the switch SW210 is turned off, and the process of fig. 5 is ended.

In step 511, since the second battery can be used, the switch SW205 is turned off and the switch SW210 is turned on to preferentially use the second battery, and the process of fig. 5 is ended.

Next, a method of estimating the OCV and resistance of each battery and the resistance of the switch SW will be described.

Here, there are a method of estimating on the device side and a method of embedding characteristic data in advance as a table, and the methods are described separately.

First, a method of estimating on the device side will be described. The OCV of each cell may be a value read by the voltage sensing line when the switch SW is open. When the switch SW is on, V-IR may be set based on a value V obtained by reading the alternator/auxiliary machine voltage from the voltage sensing line and based on the current I and the resistance (the sum of the dc resistance of the battery and the on-resistance of the switch SW is R) read from the ammeter. The dc resistance may be | Δ V/I | based on the difference Δ V between the measured voltages of the battery when the switch SW was turned on or off last time and the current I when the switch SW was turned on. The on-resistance of the switch SW may be determined in the same manner as the dc resistance from the difference between the measured voltage of the battery and the alternator/auxiliary device voltage when the switch SW is turned on or off last time. If this method is used, presetting of the battery is not necessary, and it is possible to cope with even replacement of the battery.

Next, a method of embedding the characteristic data in advance as a table will be described. OCV is represented by the sum of OCV after a sufficient time has elapsed (stable OCV) and transient voltage change (referred to as polarization) that changes in the order of several seconds. Since the stable OCV is generally expressed as a function of the charging rate of the battery, the table of fig. 7 may be held, and the stable OCV may be obtained by interpolating the table of fig. 7 based on the SOC value (fig. 7 is an example of a virtual battery). Here, the charging rate is expressed as soc (state of charge). The method for solving SOC may be as described in the literature "zui hui-wan tailang: カルマンフィルタ a kalman filter used in "upper generation publication, first generation second generation in 3.10.2013" by imperial china company, may be configured to determine the SOC in the reverse direction from the voltage of the battery at the moment ignition is turned on, as an initial value, based on the table in fig. 7, and update the SOC at every moment as a value of 100 × current integration/the capacity of the battery (current integration method). The SOC may be determined in reverse from the table of fig. 7, assuming that the measured voltage-direct current resistance × current-polarization voltage is a stable OCV (discharge is defined as + with respect to current, referred to as a voltage estimation method). Further, a weighted average may be taken for the current integration method and the voltage estimation method.

Next, a method of estimating the polarization voltage will be described. Since the charge/discharge rate in the micro-HEV is less than about 1 minute, the equivalent circuit of the battery is assumed to be fig. 8. Here, the voltage of the polarization capacitor 81 and the polarization resistor 82 corresponds to the polarization. Here, if the values of the polarization capacitance c and the resistance r are determined, the polarization voltage is determined by equation 6.

Polarization voltage i (t) × exp (-t/cr)/cr … (formula 6)

I (t): measured battery current (charge direction is set as +)

*: convolution integral

In addition, equation 7 may be used in a manner that simplifies equation 6. When equation 7 is used, vp (0) may be equal to 0.

vp (t) ═ vp (t- Δ t) (1- Δ t/cr) + i (t) × Δ t/c … (formula 7)

Δ t: time step of current measurement

vp is as follows: polarization voltage

Since the values of c and r are necessary as described above, a table of fig. 9 may be prepared. Specifically, the table of fig. 9 is interpolated from the SOC value to obtain the values of c and r. In addition, c and r may vary depending on the temperature. In this case, a thermometer may be attached to each battery, a table of fig. 9 may be prepared for each temperature, and a value may be obtained by interpolation using the measured temperature.

Next, the values of the dc resistance and the switch SW resistance will be described. First, an example of a table of dc resistance is described in fig. 10. Similarly, a value may be obtained by interpolation in the table of fig. 10 based on the SOC. Further, the dc resistance may change in value depending on the temperature. In this case, a thermometer may be attached to each battery, the table of fig. 10 may be prepared for each temperature, and the value may be obtained by interpolation using the measured temperature. The switch SW resistance may be stored in one controller 200, or may be interpolated by the controller's thermometer while holding the table of each temperature, fig. 11. Here, the dc resistance and the switch SW resistance may be values estimated from the measured values.

In the above processing, when the switch SW is switched, the battery is sometimes not connected instantaneously depending on the switching timing of the switch SW. To cope with this, the power supply system of fig. 2 may be configured as shown in fig. 12 in which the capacitor 121 is incorporated. In addition, the switch control unit 212 may be provided with a circuit for preventing both batteries from being turned off when the batteries are switched on only one side. Fig. 13 shows an example of the logic gate circuit.

In fig. 13, the switch SW205 gate signal 1311 and the switch SW210 gate signal 1312 are generated by inputting the switch SW205 signal 1301 and the switch SW210 signal 1302 a. Here, the switch SW205 signal and the switch SW210 signal refer to SW signals (TTL, Transistor and Transistor Logic signals may be used) in fig. 3, 4, and 5. Then, the time of 1 is defined as on, and the time of 0 is defined as off. In addition, the switch SW205 gating signal refers to a signal line of the switch SW205 in fig. 2, and the switch SW210 gating signal refers to a signal line of the switch SW210 in fig. 2. Here, fig. 13 corresponds to the SW control unit 212 in fig. 2. First, in order to prevent the double-break signal from being generated, the condition for double-break is determined by the OR logic gate 1302b and the NOT logic gate. The signal that turns both off is 1304. When the both-sides disconnection is performed, the first battery is forcibly connected to the safe station (site), and therefore, the OR logic gate 1309 sets the switch SW205 gating signal as the candidate of the switch SW205 signal and the OR of the both-sides disconnection signal. During parking, the controller 200 is in the sleep state, and thus the power supply to each logic gate is turned off. In this case, since the first battery is used as a power source of the protection device, the pull-up resistor 1310 is used to forcibly set the signal to 1, and the switch SW205 is kept on. In the case where a mechanical relay is used for the switch SW205, the switch SW terminal may be connected so that the relay is turned on at a position where the logic gate current (electromagnet of the relay) becomes 0, and therefore, a pull-up resistor is not required. Further, even in the case of using a relay of a closed lock type (a type that maintains the previous state of the switch SW even if the electromagnet current does not flow) among the mechanical relays, the present invention is not limited thereto (the pull-up resistor may not be required and the switch SW terminal in a state in which the current 0 is considered may be not considered). If a FET or an IGBT is used for the switch SW205, a FET (or IGBT) driver may be added. In this case, the FET driver also does not disconnect the power supply during parking. If the current consumption of the FET driver is large, the mechanical relay may be connected in parallel only to the switch SW205, and turned on if the current of the electromagnet is 0, so that the FET shares the operation of the switch SW during the ignition on process, thereby extending the life of the mechanical relay. Alternatively, a diode (the current direction of the diode is from the first battery to the alternator/accessory side) may be connected in parallel to the switch SW205 without using a pull-up resistor. In addition, when the FET is used for the switch SW210, it is preferable to turn off the switch during parking, and therefore, a pull-down resistor may be incorporated in the logic gate of the switch SW 210.

Next, the delay is explained. Here, when the signal switches SW205 and SW210 are turned off from on, there is a possibility that both are turned off due to a delay until the switch SW is turned on and a delay of the logic gate. Therefore, by providing circuits (1306, 1307) for delaying the switch SW signal, the delayed signal and the original signal are OR (OR logic gates 1305, 1308) to prevent both from being disconnected. In addition, in order to prevent the switch SW205 signal from being turned off by the interruption of the program, the signal is set as a candidate for the switch SW205 signal, and an OR with the both-off state signal is set as the switch SW205 gate signal 1311 as described above. As for the gate signal of the switch SW210, a signal of an OR logic gate 1308 is used. The delay circuit may be a circuit in which an integrating circuit is connected to a gate signal and an output of the integrating circuit is received by a schmitt trigger. Here, the delay time may be determined as a predetermined value based on the on delay time of the switch SW + the delay time of the logic gate.

Fig. 13 is a circuit, but may be a logic of a program equivalent to fig. 13 (when a pull-up resistor is put in, a pull-up resistor is added to an I/O signal of the CPU serving as a gate signal of the switch SW 205).

In addition, although the pull-up resistor and the pull-down resistor described above may be changed in the case where the FET is connected to GND or the positive electrode voltage of the first battery in the P channel or the N channel, the FET may be connected so that the switch SW205 is turned on and the switch SW210 is turned off during parking.

Next, a description will be given of a countermeasure in a case where the voltage of the first battery or the second battery is reduced by self-discharge during long-term parking. The processing in this case will be described as an example of processing included in the ignition on time and processing 34 in the controller wake-up in fig. 3. An example of the processing in this case will be described as fig. 14. Fig. 14 is a process after the controller 200 wakes up.

In step 141 after the wake-up, it is determined whether the first battery voltage (which coincides with the OCV) is less than a certain threshold. If the value is smaller than the threshold value, the process proceeds to step 143, and if the value is greater than or equal to the threshold value, the process proceeds to step 142. The threshold value may be set in advance to a value corresponding to a voltage of the SOC 80% or 90%. In this case, a table (fig. 7) of the charging rate and OCV of the first battery needs to be held in advance.

In step 143, since the charging rate of the first battery is insufficient, the switch SW205 is turned on and the switch SW210 is turned off, and the process proceeds to step 144. In step 144, a start instruction is output to start charging of the first battery, and the process proceeds to step 145. Here, the driver may wait until the engine is started without sending the start command.

In step 145, the process is repeated until the charging rate of the first battery becomes equal to or higher than the charging rate of the first battery, and if the charging rate becomes equal to or higher than the threshold value, the process proceeds to step 146. The procedure of the method of solving the state of charge here will be described. First, charging is started, and a function approximation is performed by taking the current time series i (t) after the current becomes CV charging as an exponential function (coefficients x, y, z of equation 8 are calculated). Note that R1 may be calculated as R1 (voltage immediately after charging-voltage immediately before charging)/(current immediately after charging-current immediately before charging) using the current and voltage before and after the start of charging, or may be calculated using values previously stored in the table as described above.

V (t) first cell voltage measurement-first cell current measurement xr 1-initial OCV

… (formula 8)

x-polarized resistance x polarized capacitance

y polarization resistance x (1+ polarization capacitance/capacity of first battery)

z is 1/capacity of first battery [ F ]

Here, since the time series of q (T), f (T), g (T) can be obtained from the measured values, if there is accumulation of data at the time τ, time τ + Δ T, and time τ + (n-1) Δ T at which CV charging is started, x, y, and z can be obtained as expression 9 by the least square method.

Whether or not the state of charge is a predetermined value may be determined as step 146 when equation 10 is satisfied, using z obtained by equation 9. Equation 9 may be required to store the past time series, and may not be calculated according to the CPU specification of the controller 200. In this case, a recursive least squares method may also be used, which updates the time series one by one (goodness, autumn, middle channel, mountain Paisha: システ harmony, rotation of the government, 1981). As the target charging rate, the value of the charging rate (e.g., 80%, 90%) used in step 141 may be used.

Target charge rate OCV-initial OCV ≧ Q (t) z. (equation 10)

Next, at step 146, it is determined whether or not the voltage of the second battery is equal to or higher than the threshold value, and if it is lower than the threshold value, the process proceeds to step 147, the switch SW205 is turned off, the switch SW210 is turned on, and the process is repeated until the voltage of the second battery becomes the threshold value. When the voltage of the second battery becomes equal to or higher than the threshold value, the process proceeds to step 148, and an instruction to stop the engine is sent to the ECU 15. If idling is stopped, the AC generator may be stopped on the ECU side. Then, the flow of fig. 14 is ended. Here, the threshold value is the lowest voltage of the second battery.

At step 142, it is determined whether or not the voltage of the second battery is smaller than the threshold value, and if so, the second battery needs to be charged, so the process proceeds to step 143. If the value is equal to or greater than the threshold value, the process of FIG. 14 is terminated. As the threshold value here, the same value as the threshold value explained in step 146 is used.

As described above, since the charging current of the battery system can be estimated during regenerative charging, the estimated charging current is transmitted to ECU 15. Since the torque of the ac generator changes (is proportional to the current) on the ECU side, mechanical brake emphasis control may be performed to prevent torque discontinuity during braking and optimize ride quality. Instead of the charging current, the charging power may be transmitted to ECU 15. The charging power is calculated as the estimated charging current × alternator/auxiliary machine voltage.

Next, embodiment 1 (36 in fig. 3) in which switching of the switch is performed only once during a period from the start to the end of regeneration will be described. By prohibiting the parallel connection of the batteries and performing the switching of the switch SW only once, the cross current can be suppressed as much as possible, and the noise caused by the switching of the switch SW can be suppressed.

As the timing of switching, for example, a current time series in the case where only one cell is connected is estimated, and when the current time series of the first cell is represented by I (T) and the current time series of the second cell is represented by I (T), the time series I (T- τ) and I (τ) are compared and are made equal to each other, the solution τ of the nonlinear equation is obtained, and when τ has elapsed from regeneration, the second cell is switched to the first cell (T is the regeneration time).

Fig. 15 shows an example of the process of the regenerative charging control. In step 151, the second battery is first selected (the first battery may be first charged, an example of which will be described later). Specifically, the switch SW205 is turned off, and the switch SW210 is turned on.

Next, in step 152, the estimated current time series i1(t) in the case where only the first cell is selected is received. The current estimation unit 204 in fig. 2 calculates the estimated current time series.

Next, in step 153, the estimated current time series i2(t) in the case where only the second battery is selected is received. The current estimation unit 209 in fig. 2 calculates the estimated current time series.

Here, the calculation method of the estimated current time series and the reception format example in

steps

152 and 153 will be described. Here, the format examples include two types, i.e., a case where a formula is assumed in advance and a constant in the transmission formula and an example of a transmission current time series. The description is given separately.

First, three kinds of batteries, i.e., a battery (large-capacity battery) having a large capacity and an OCV that hardly changes during primary regeneration, such as a lead battery, a battery (small-capacity battery) having a small capacity and an OCV that greatly changes during primary regeneration, and a battery (not precisely, a battery, but expressed herein as a battery) having negligible polarization, such as an accumulator, will be described. Fig. 16 is assumed for an equivalent circuit of a large-capacity battery. In this case, by solving the circuit equation of fig. 16, the current time series i (t) of equation 11 is obtained.

Ia: maximum current capable of generating power by AC generator-auxiliary machine current

1 ∞ ═ (Va-VL)/(RL + RL): stabilizing the current

VL: OCV of large capacity battery

Va: CV voltage of AC generator

RL: DC resistance + switch SW resistance

rL: polarization resistor

τ L: time constant RL rL cL/(RL + rL)

cL: polarization capacitor

Kappa: CC charge end time ═ cL × rL { (Ia × rL-vp (0))/(VL + (rL + rL) Ia-Va) }

vp (O): initial value of polarization voltage

The equivalent circuit of the small-capacity battery is approximated to fig. 17. This is the case when the function of ocv (soc) is approximated to be linear. In this case, the current time series of equation 12 is obtained by solving the circuit equation of fig. 17.

C: capacitance [ F ] representing the slope of the OCV (SOC) curve

Rs: DC resistance + switch SW resistance

cs: polarization capacitor

rs: polarization resistor

vs: polarization voltage at the end of CC charging is Va-Vs-Ia Rs-1a kappa/C

λ 1, λ 2: solution of the equation 2(λ 1 > λ 2)

λ1+λ2＝C(Rs+rs)+cs*rs，λ1*λ2＝C*Rs*cs*rs

k：

W (x): mingber W function

Va: CV voltage of AC generator

vs (O): initial polarization voltage

Vs: initial constant OCV

Fig. 18 shows an equivalent circuit of the capacitor. In this case, the time series of equation 13 is obtained by solving the circuit equation of fig. 18.

Rc: DC resistance + switch SW resistance

Cc: electrostatic capacity of accumulator

κ：

Vc (O): initial OCV of an electrical storage device

Therefore, the functions of equation 11, equation 12, equation 13, and the like may be determined in advance by the first battery, the second battery, and the type of each battery, and the coefficients may be calculated and transmitted. Here, instead of the coefficients, the current time series in the CC charging period, Ia, CV charging period may be transmitted. The time series may be 1s or 0.5 s. The CC charge time κ of the small-capacity battery is a lambertian W Function, and a numerical calculation method (refer to Chapeau-Blondeau, f.and pair, a: Evaluation of the lambertian W Function and Application to Generation of Generalized Gaussian Noise With exposure 1/2, IEEE trans. signal Processing, 50(9), 2002) With high speed and good accuracy is proposed for the lambertian W Function, and therefore can be calculated by the CPU of the controller 200. Alternatively, a table of the W function may be set in advance, and the value may be obtained by interpolation. Next, a method for solving the coefficients in

equations

11, 12, and 13 will be described. The dc resistance and the switch SW resistance may be obtained from the ratio of the past current change and voltage change as described above, or values set in advance in a table may be used. The initial OCV may be a measured voltage value — a polarization voltage when the switch SW connected to each battery is off. What becomes problematic here is the polarization voltage. When the polarization capacitance and the polarization resistance are determined, the polarization voltage may be calculated by measuring the current and using equation 6 or equation 7. The polarization capacitance and the polarization resistance may be obtained by interpolation while holding a table as described above (but preparing a table for charging separately), or may be obtained by time series data in the past. As this method, equation 9 may be obtained using the measurement data at the previous CV charge (however, V and I may be obtained from the measurement values at the time of charge, and Q, f, and g may be obtained). In equation 9, the polarization capacitance and the resistance are obtained by solving simultaneous equations based on x-polarization capacitance × polarization resistance, y-polarization resistance x (1+ polarization capacitance/capacitance C), and z-1/capacitance C. In the case where the processing of fig. 14 is not performed at the time of the first ignition on, the polarization capacitance and the resistance are unclear, and thus, values immediately before the end of the previous travel may be used. The CV charging of the battery may be performed as a set value at factory shipment, and an initial value may be set based on measurement data. Ia may be received from ECU 15.

Next, in step 154, the switching time τ is determined so that the charge becomes maximum. This notion is described. If the current time series of the first battery is i1(T), the current time series of the second battery is i2(T), and the regeneration time is T, the charge Q during regeneration is equation 14.

Since the maximum of equation 14 is τ where dQ (τ)/d τ is 0, the solution τ of equation 15 is the switching time when the charge is maximum.

i2(τ) ═ i1(T τ) … equation 15

Equation 15 is a non-linear equation. Fig. 19 shows an image of the solution. Here, i2 is 191 (corresponding to i2(τ)), and i1 is drawn with the time axis reversed (corresponding to i1 (T- τ)) is 192. The intersection 193 is the solution τ of equation 15. Therefore, the current at the switching time τ of the second battery and the current at the end of regeneration of the first battery coincide with each other.

That is, when the regeneration time is T, the charging time of the second battery is τ (T > τ), and the first charging time is T- τ, the switch SW is switched at the timing τ at which the amount of charge of the first battery and the second battery becomes maximum (the first battery and the second battery may be reversed). τ is a timing at which the total value of the value obtained by integrating the equation describing the charging time and the current value of the second battery from the charging time 0 to τ and the value obtained by integrating the equation describing the charging time and the current value in reverse in the range from T to τ of the first battery becomes maximum during the regeneration time T.

Non-linear equation of formula 15 can be solved by Newton's method or by dichotomy (refer to literature, Mitsui nationality; number value Algorithm 2 nd edition, published by North-son, 2014). In addition, the above formula may be used to calculate the value of the current function in the middle of the numerical calculation, and when it is assumed that the current function is provided by time series data, the value of the current function may be calculated by interpolation.

In this case, in fact, with respect to i1, the absolute value of the polarization voltage decreases due to τ, and thus there is also a possibility of deviation from the estimated current time series. As this correction, a function of expression 16 may be used as the correction of κ of expression 11, and a function of expression 17 may be used as the correction of expression 12.

Equation 15 represents the case where the regeneration completion time T is determined. T is good when information is obtained from the ECU (in this case, T may be obtained from the speed and deceleration on the ECU15 side, for example), but it is not always clear. When T is not clear, since the first battery uses a large-capacity battery, I1 may be calculated as I ∞ or may be prepared in advance for a typical time (e.g., 5s or 10s) required for regeneration.

After τ is obtained in step 154, the process proceeds to step 155, and it is determined whether or not the elapsed time from the start of reproduction is τ or more. If it is above τ, the process is transferred to step 157, otherwise the process is transferred to step 156. It is determined in step 156 whether or not the regeneration is ended, and if the regeneration is not ended, the process returns to step 155. The determination as to whether or not the regeneration is completed is determined based on whether or not the signal from ECU15 or the sum of the ammeters has a value equal to or greater than 0. If the regeneration is finished, the processing of FIG. 15 is finished.

In step 157, a selection instruction to the second battery is sent to the switch SW. Specifically, the switch SW205 is turned on, and the switch SW210 is turned off. After the switch SW is switched, the process is transferred to step 158. When the regeneration is completed, step 158 ends the processing of fig. 15.

In the above processing, the second battery is turned on first, but the first battery may be turned on first.

Example 2

Referring to fig. 20, the flow of embodiment 2 of the micro HEV of the present invention will be described in detail. Example 2 is an example in which the switch is switched a plurality of times during a period from the start to the end of regeneration. The estimated currents of the first battery and the second battery are obtained at constant intervals, and the switch is switched to the larger one of the estimated currents of the first battery and the second battery for each charging.

Specifically, the switch is periodically switched when both the first battery and the second battery are CC chargeable. After one of the CC charges is completed, the other battery is charged, and then the switch is switched to the one of the estimated currents of the first battery and the second battery, which has the larger estimated current, to perform the charging. The details are as follows.

This example shows the processing in the following case: in order to prevent as much as possible a cross current during charging during regeneration (which may occur instantaneously at the instant when the switch SW is switched), the batteries are inhibited from being connected in parallel, and switching of the switch SW is permitted a plurality of times in 1 regeneration (the rest of the processing is the same as in embodiment 1). That is, fig. 20 is implemented without implementing fig. 15 in embodiment 1.

First, in step 2001, the estimated current i1 immediately after connecting only the first battery is received. This calculation is performed in the current estimation unit 204 of fig. 2. As a calculation method thereof, when the switch SW205 is currently off, (CV voltage of the ac generator — measured voltage of the first battery)/(direct current resistance of the first battery + on resistance of the switch SW 205) is set. As for the direct current resistance and the on-resistance of the switch SW, the method described in embodiment 1 is used. When the current switch SW205 is on, the measured current value is set. Thereafter, the process shifts to step 2002.

In step 2002, the estimated current i2 immediately after connecting only the second battery is received. This calculation is performed by the current estimation unit 209 in fig. 2 (hereinafter, regarding the current, the charging direction is defined as +). As a calculation method thereof, when the switch SW210 is currently off, it is assumed that (CV voltage of the ac generator — measured voltage of the second battery)/(direct current resistance of the second battery + on resistance of the switch SW 210). As for the direct current resistance and the on-resistance of the switch SW, the method described in embodiment 1 is used. When the current switch SW210 is on, the measured current value is set. After that, the process shifts to step 2003.

In step 2003, it is determined whether both batteries become CC charged (constant current charging), and if both batteries are CC charged, the process is shifted to step 2008, otherwise the process is shifted to step 2004. As a determination method, Ia (generated current of ac generator — auxiliary machine current) described in example 1 was used, and i1 ≧ Ia and i2 ≧ Ia were set.

In step 2004, i1 and i2 are compared, and if i1 is large, the process is transferred to step 2005, otherwise the process is transferred to step 2006. In step 2005, a switch SW instruction to select the first battery is sent. Specifically, the switch SW205 is turned on, and the switch SW210 is turned off. Thereafter, the process is transferred to step 2007. In step 2006, a switch SW instruction is sent to select the second battery. Specifically, the switch SW205 is turned off, and the switch SW210 is turned on. Thereafter, the process is transferred to step 2007. The reason for this is that by selecting a battery with a large charging current, the charging charge is increased.

In step 2008, since the CC is charged regardless of which battery is connected, a process "CC charge time SW selection process" is executed to extend the CC charge time. This processing will be described later. After step 2008 ends, the process is transferred to step 2007.

In step 2007, it is determined whether or not the reproduction is ended, and if not ended, the process proceeds to step 2001, and if ended, the process of fig. 20 is ended. Note that these processes may be returned to the process of step 2001 every measurement cycle (for example, every 10ms) or every 0.1 s.

As the process of selecting the switch SW during CC charging in step 2008, the duty ratio of the time when the first battery is selected and the time when the second battery is selected may be represented by η: 1-eta (0. ltoreq. eta. ltoreq.1). The switch SW may be switched periodically. Here, η may be a predetermined value, the method of fig. 21 may be used, or η may be changed according to the measured voltage or current.

Next, the CC charging time switch SW selection process of fig. 20 will be described with reference to fig. 21. This process is intended to alternately switch the batteries to allow one battery to rest, reduce the polarization voltage, increase the next charging current, prolong the overall CC charging time, and increase the charging charge during regeneration.

First, in step 2101, it is determined whether the estimated current of the first battery is smaller than the CC charging current Ia, and if smaller than Ia, the process is shifted to step 2102, otherwise the process is shifted to step 2103.

In step 2102, an instruction to select the switch SW is sent to the first battery. Specifically, the switch SW205 is turned on, and the switch SW210 is turned off. After that, the process of step fig. 21 is ended (the process is shifted to step 2007 in fig. 20).

In step 2103, an instruction to select the switch SW is sent to the second battery. Specifically, the switch SW205 is turned off, and the switch SW210 is turned on. After that, the process of step fig. 21 is ended (the process is shifted to step 2007 in fig. 20).

In the loop of the processing in fig. 20 and 21, the CC charging of the first battery is first completed. Then, the second battery is set to CC charge at Δ t (equivalent to resting the first battery). During this period, the polarization voltage of the first cell decreases by Δ t × vp/cLrL (vp is the polarization voltage at the end of CC charge, cL is the polarization capacitance, and rL is the polarization resistance). Therefore, when the first battery is switched, CC charging can be performed by Δ t × vp/(Ia × rL-vp). Therefore, the CC charging time of the first battery is further extended, and the charge is increased. In addition, regarding the duty ratio, the ratio of the on-time of the first battery to the on-time of the second battery becomes vp: ia r l-vp. In the following, the reason why the CC charging of the first battery is prioritized first will be described.

The CC time κ due to the switching of the battery is frequently switched by the switch SW, and if the time rate of turning on the first battery is η, the time of the first end of CC charging of the first battery is λ, the equivalent circuit of the first battery is approximated to fig. 16, and the equivalent circuit of the second battery is approximated to fig. 18 regardless of the polarization of the first battery, the CC time κ can be approximated by equation 18 by solving the circuit equation.

According to equation 18, to maximize κ, it is equivalent to maximizing equation 19.

Equation 19 is equivalent to maximizing equation 20 according to the circuit equation of fig. 16.

vL (t): polarization voltage of first battery (< vp)

To maximize equation 20, vL may be made to approach vp the fastest, according to vL-vp < 0. Therefore, initially, in order to make the polarization voltage of the first battery most close to vp, the first battery is connected.

Here, if the duty ratio of the first battery after the first battery becomes vp is η, η is vp/Ia rL, and therefore, the polarization resistance rL may be obtained from a result obtained by observing the duty ratio η by the controller 200, and the values (vp is Va-VL-Ia rL, and rL is vp/(Ia η)) may be held.

Next, a case where the duty ratio η is changed in accordance with the measured current voltage (basically, a case where a small-capacity battery is used as the second battery) will be described. Here, two ideas of controlling the polarization voltage of the large-capacity battery (first battery) and controlling the OCV of the small-capacity battery (corresponding to the voltage-IaRs of the small-capacity battery by the voltage 171 in fig. 17 of the second battery) will be described.

In the former case, η may be determined as equation 21 with the voltage time increase rate of the first battery set to m1 (where a limit is added to the range of 0 ≦ η ≦ 1). Equation 21 is derived from the equation of the circuit assuming that a current of η × Ia substantially flows in the polarization.

vp (t) polarization voltage of first cell-first cell terminal voltage-Ia RLVL

In the latter case, the voltage time increase rate of the small-capacity battery may be set to m2 and η may be determined as equation 22 (with the addition of a limit set to a range of 0 ≦ η ≦ 1). This is derived from the circuit equation assuming that the current of (1- η) Ia substantially flows in the polarization of the small-capacity battery (second battery).

vs (t): polarization voltage of the second cell-1 a Rs-stable OCV of the second cell

C: electrostatic capacitance [ F ] corresponding to the Slope of OCV (SOC) in FIG. 17

Here, when it is desired to match the end time of CC charge of the first battery and the end time of CC charge of the second battery from time t1, the relationship between m1 and m2 in equation 23 may be used.

Vs (t 1): OCV (including polarization) of second cell at time t1

The denominator of equation 23 is "OCV at the time point when CC charging is completed — OCV at the current time point". Here, Vs (t1) may be a voltage at the time when the switch SW210 of the second battery is off, or may be a terminal voltage-IaRa of the second battery when the switch SW210 is on.

Here, if a (t) is defined by expression 24 based on expression 23, then m1 and m2 are expressions 25.

Since m1 and m2 are preferably non-negative values and as small as possible, η (where vp (t)) may be set to m1 ═ 0 in formula 21 after a (t) ∞ (i.e., after vp (t) ═ vp) or Ia ═ vs (t)/(rs × (1+ cs/C)) + vp (t))/rL has been established (where vp (t)) needs to be non-negative). The condition of vp (t) is the same as the above-described case where only the first battery is connected and CC charging is completed.

Here, after Ia ═ vs (t))/(rs × (1+ cs/C)) + vp (t))/rL is established, the CC charge time by switching can be extended indefinitely (actually, the OCV of the large-capacity battery changes, and therefore, the CC charge time becomes finite, but if the regeneration time is, for example, about 10s, the CC charge time is sufficiently extended, and the charge becomes maximum. Therefore, η may be controlled so that Ia ═ vs (t)/(rs × (1+ cs/C)) + vp (t))/rL is first set. As a control method, η is determined so as to minimize the control target in order to reduce both voltage rises according to the circuit equation of equation 26. The control targets are:

(dv_p(t)/dt)²+const.(dV_s(t)/dt+dv_s(t)/dt)²＝(I_oη/c_L-v_p/c_Lr_L)²+const.

(-I_oη/c_s+I_a(1/c_s+1/C)-v_s/c_sr_s)²∝(η-vp/I_ar_L)²+const·((1+c_s/C)(η-1)+v_s/c_sr_s)²

modern control theory can also be used as a control unit for determining this η.

The equation of state:

observation equation: v₁(t)＝V_L+I_aR_L+v_p(t)

V₂(t)＝V_s(t)+I_aR_s+v_s(t)

V₁′(t)＝V_L+v_p(t)

V₂′(t)＝V_s(t)+v_s(t)

V1 (t): first battery measures voltage (when switch SW205 is on)

V2 (t): second Battery measuring Voltage (when switch SW210 is on)

V1' (t): first battery measures voltage (when switch SW205 is open)

V2' (t): second Battery measuring Voltage (when switch SW210 is OFF)

The restriction conditions are as follows: eta is more than or equal to 0 and less than or equal to 1

v_p(t)≤V_a-I_aR_L

v_s(t)+V_s(t)≤V_a-I_aR_s

After the condition of Ia ═ vs (t)/(rs × (1+ cs/C)) + vp (t)/rL is satisfied, η (η ═ vp (t)/Ia ×) set to m1 ═ 0 is fixed in formula 21. In addition, as long as the controllable condition is not satisfied, Ia ═ vs (t))/(rs × (1+ cs/C)) + vp (t))/rL may not be true anyway. For example, the case is described below in which Ia > vsm/(rs × (1+ cs/C)) + vp/rL (vsm is the upper limit of the polarization voltage of the small-capacity cell, Ia × rs (1-exp (- κ/(cs × rs)) + vs (0) exp (- κ/(cs × rs)), and κ is κ or cl ═ rL ═ cs × rs of formula 12, first after vp t ═ vp, and η ═ vp/Ia ═ rL may be set as described above.

As another method of extending the CC charge time, the second battery alone may be CC-charged first, and if the second battery alone cannot be CC-charged, the switching control of the switch SW may be performed by determining the duty ratio as η when m2 is equal to 0 in equation 22.

Example 3

Referring to fig. 22, the flow of embodiment 3 of the micro HEV of the present invention will be described in detail. The third embodiment is an example in which the switch is switched a plurality of times during the period from the start to the end of regeneration, and both batteries may be connected so as to be charged simultaneously. After the CC charging of the second embodiment is completed, the switches of the 2 batteries are turned on. In this case, by combining the constant monitoring of the cross current and immediately combining the control such as turning on the switch SW of only 1 cell having a larger charging current when the cross current is generated, the cross current can be prevented.

This example is a case where simultaneous connection of batteries is permitted in order to increase the charge during regeneration at the time of charging during regeneration (the remaining processing is the same as in embodiment 1). That is, fig. 22 is implemented without implementing fig. 15 in embodiment 1.

First, in step 2201, a charging current i1 in the case where only the first battery is connected is estimated. This is the same as the method described in example 2.

Next, at step 2202, a charging current i2 in the case where only the second battery is connected is estimated. This is the same as the method described in example 2.

Next, at step 2203, the charging current of each battery when both the first battery and the second battery are connected is estimated (I1 for the first battery current and I2 for the second battery current). The method is described. The equivalent circuit in the case of parallel connection is fig. 6. In fig. 6, since the discharge direction is set to + the sign of the current is reversed, I1 and I2 (in CC charge) can be calculated. However, in CV charging, values of I1 and I2 are used because I1 is I1 and I2 is I2. As the determination of CV charging, if the voltage V in fig. 6 is Va (CV voltage of the ac generator) or more, it is determined that CV charging. Here, the dc resistance, the on-resistance of the switch SW, and the OCV of each battery are estimated by the method described in example 1. Here, when both the switches SW are currently on, the measured current may be used.

Next, in step 2204, it is determined whether both i1 and i2 become the CC charging current Ia or more (whether both become the CC charging in the case of using only one of the first battery and the second battery). If either battery becomes CC charged, the process is transferred to step 2205, otherwise the process is transferred to step 2206.

In step 2205, a switch SW selection process during CC charging (the same process as the process in fig. 21 during CC charging described in embodiment 2) is executed, and the process proceeds to step 2207.

When connected simultaneously in step 2206, it is determined whether cross flow is generated, and if so, the process is transferred to step 2208, otherwise, the process is transferred to 2210. As this determination method, I1<0 or I2<0 is set.

In step 2208, since the cross current is generated at the time of simultaneous connection, it is determined which cell is selected. In the case where the estimated current of the first battery > the estimated current of the second battery, the process is transferred to step 2209, otherwise the process is transferred to step 2211.

In step 2209, a command for selecting the first battery is transmitted (switch SW205 is turned on, and switch SW210 is turned off), and the process proceeds to step 2207.

In step 2211, a command to select the second battery is transmitted (switch SW205 is turned off, and switch SW210 is turned on), and the process proceeds to step 2207.

At step 2210, a command for selecting both batteries is transmitted (switch SW205 and switch SW210 are turned on), and the process proceeds to step 2207.

It is determined in step 2007 whether or not the regeneration is ended, and if not ended, the process proceeds to step 2201, otherwise (if ended), the process of fig. 22 is ended.

In the above description, since I1+ I2> I1 and I2 are generally used, the comparison process of I1, I2 and I1+ I2 is omitted, but the comparison process may be introduced.

In the above case, when CC charging cannot be continued only in a single battery, CC charging can be continued by connecting both batteries, and therefore, the charge increases.

Next, in step 2205, a process for allowing simultaneous turn-on may be introduced during the CC charging period. The method is described. Time set to connect only the first battery: time to connect only the second battery: time for connecting both batteries ═ η: ζ: 1- η - ζ (0< η, ζ < 1). η, ζ may be controlled by a value provided in advance.

Next, in the case where the parameters (C, cs, rs) of the small-capacity battery are not set in the table, it is necessary to identify the parameters (the direct current resistance can be identified by the above method, and the case of the large-capacity battery is described in equation 9). The identification method is described. In the case of the electric storage device, the same as the case where Rs is 0, C is Cs, and Rc is Rs, and therefore, the case where the electric storage device is used in the second battery is omitted. The identification timing may be two types, i.e., a CC charge case and a CV charge case. However, since control is also introduced at the time of CC charging, there is no interaction between batteries even if the batteries are connected in parallel to a constant voltage source at the time of CV charging, and thus parameter identification at the time of CV charging is described here.

In the case of CV charging of a small-capacity battery, the equation of equation 27 is given.

1 (t): measuring current

Therefore, the linear equation of equation 28 holds for the ambiguous parameters C, rs, and cs.

xi (t) + yq (t) + zf (t) ═ g (t) … (formula 28)

G(t)＝(V_a-V_s(0))t-R_sQ(t)

x＝c_sr_sR_s

In equation 28, x, y, and z may be obtained by the least square method or the recursive least square method, and C, cs, and rs may be obtained from x, y, and z. In the case of an electric storage device, C may be obtained by using the least square method or the recursive least square method in the same manner, assuming that x is 0.

Next, a method for solving the ambiguous parameters VL, cL, and rL of the large-capacity battery will be described. With respect to the difference from the small-capacity battery, g (t) in formula 28 is VL t-RL q (t), and C is infinity, i.e., z is 0 and y is rs, to obtain formula 29.

xI(t)+r_sQ(t)-V_Lt＝-R_sQ (t) … (formula 29)

In equation 29, x, rs, and VL may be obtained by the least square method or the recursive least square method, and cs may be obtained from x. In addition, although the above is the case of charging, the battery parameters of charging and discharging may be different depending on the battery. In this case, the battery parameters may be identified similarly using the data at the time of discharge.

Finally, comparison of the effects of the charge in the regenerative charging described above will be described. Here, an example is given in which a lead battery is used for the first battery and a battery (lithium ion battery) is used for the second battery, and the respective factors are assumed to be fig. 23. Then, in the ac generator, the voltage at the time of CC charging 200A, CV charging is assumed to be 14V, and the switching resistance is assumed to be 0 in case of regeneration for 10 s. In addition, lead batteries alone, accumulators alone, lead batteries and accumulators in parallel at all times (i.e., the initial OCV of the lithium ion accumulator was 12.6V, which is the same as that of the lead batteries) were compared, the method of switching the switches SW once described in example 1, the method of non-simultaneous connection described in example 2 (the method in which the lead batteries were charged first and η ═ vp/Ia rL was set at the time point when the polarization of the lead batteries became vp, the method of simultaneous connection described in example 3 (the method in which the lead batteries were charged first and switched to η ═ vp/Ia rL at the time point when the polarization of the lead batteries became vp and the method of simultaneous connection after CC charging was not possible only for one battery) described in example 3 were compared. Fig. 24 shows the comparison result obtained by numerical calculation with a time step of 50 ms. According to fig. 24, the proposed embodiments (example 1, example 2, and example 3) have a larger charge than the case where the batteries are connected in parallel at all times by themselves. In addition, in example 3, the charge was the most charged. This is because the CC charging time becomes long because simultaneous connection is allowed. The reason why the charge is low when the capacitors are always connected in parallel is that the initial voltage of the capacitors is as low as 12.6V, and the charge that can be charged in the capacitors is small.

Example 4

The flow of embodiment 4 in the micro HEV of the present invention will be described. In the present invention, one battery is first charged to a voltage at which no cross current is generated, and then connected in parallel. Since the CC time is not increased by the decrease in the polarization voltage as compared with embodiment 3, the charge is reduced as compared with embodiment 3, but the number of times of switching the switch SW is reduced, and the noise is suppressed.

This processing will be described according to the flow of fig. 25. Steps 251 to 253 are the same as 2201 to 2203 of embodiment 3. In step 254, it is determined whether cross flow is generated when simultaneously connected. Then, if the cross current is not generated, the process proceeds to step 255, where both switches SW are turned on. If a cross flow is generated, the process is transferred to step 256.

In step 256, an undervoltage condition of the second cell, i.e., "V1-V2 > Ia × Rc", is determined. V1 is OCV of the first battery (unstable value, i.e., one of battery voltages when the current is 0), V2 is OCV of the second battery (unstable value, i.e., one of battery voltages when the current is 0), and Rc is direct-current resistance of the second battery. If the voltage of the second battery is insufficient, then the process moves to step 257 where the second battery is selected and only the second battery is charged. Otherwise, since it is the first battery that is under-voltage, the process moves to step 258, where the first battery is selected and only the first battery is charged.

After the switch SW processing at

steps

255, 257, and 258, the processing proceeds to step 259, where it is determined whether or not the reproduction is completed. If the regeneration is not ended, the process returns to step 251, and if the regeneration is ended, the process of fig. 25 is ended.

Claims

1. A battery system in which a first battery and a second battery are connected in parallel via a switch SW,

the switches SW are a first switch SW and a second switch SW,

the first switch SW and the second switch SW are connected in parallel,

the first battery is connected to a load via the first switch SW,

the second battery is connected to the load via the second switch SW,

the battery system has a unit that estimates a charging current of the first battery from an internal resistance and an open circuit voltage OCV of the first battery and a unit that estimates a charging current of the second battery from an internal resistance and an open circuit voltage OCV of the second battery,

the switch SW is switched so that the sum of the charge charges of the first battery and the second battery becomes maximum during the regeneration time based on the estimated charge current of the first battery and the estimated charge current of the second battery.

2. The battery system according to claim 1,

the second battery is discharged first at the time of discharge, and when the second battery becomes a predetermined voltage or a predetermined state of charge, the first battery is switched to discharge, or,

the first battery is discharged first, and when the first battery reaches a predetermined voltage or a predetermined state of charge, the second battery is switched to discharge.

3. The battery system according to claim 1 or 2,

the mode is a mode in which the switch SW is switched once at the time of regenerative charging,

when the regeneration time is T, the charging time of the second battery is tau, and the first charging time is T-tau, wherein T > tau,

the charging of the second battery is performed first,

the switch SW is switched to the first battery at a timing τ at which the sum of the amounts of charge of the first battery and the second battery becomes maximum.

4. The battery system according to claim 3,

τ is a current time series (i1(t)) of the first battery and a current time series (i2(t)) of the second battery, and the switch SW is switched so that the second battery is charged at first and the first battery is charged after t has elapsed from the start of charging at a time t from the start of charging when i1(t) is i2 (charging end time-t).

5. The battery system according to claim 3,

when the regenerative charging time T is unknown, the time T at which the current timing of the first battery is obtained as the current convergence value of the second battery is taken as τ.

6. The battery system according to claim 1 or 2,

the mode is a mode in which the switch SW is switched a plurality of times during regenerative charging,

the estimated charging current of the first battery and the estimated current of the second battery are compared, and the switch SW is switched to charge the battery with a larger estimated current.

7. The battery system according to claim 6,

the switch SW is periodically switched to charge the first battery or the second battery until one of the first battery and the second battery is charged by the constant voltage.

8. The battery system according to claim 7,

when the first battery becomes a constant-current charge state, the time ratio of the switch SW of the first battery is first set to 1, and after the first battery becomes a constant-current charge end state, the switch SW time ratio of the first battery is set to (voltage at constant-voltage charge of the ac generator-open voltage of the first battery)/(current at constant-current charge of the ac generator × polarization resistance).

9. The battery system according to claim 7,

the load to which the first battery and the second battery are connected via the switch SW is an alternating current generator,

when the first battery becomes a constant current charge, the time ratio of the switch SW of the first battery is controlled so that the current when the constant current charge of the ac power generator is equal to the polarization voltage of the second battery/(polarization resistance of the second battery × (1+ polarization capacitance of the second battery/capacitance of the second battery)) + the polarization voltage vp of the first battery (t))/the polarization resistance of the first battery.

10. The battery system according to claim 1 or 2,

when the switch SW is switched a plurality of times during charging and simultaneous connection is permitted, the charging currents of the first battery and the second battery in 3 cases when the first battery alone and the second battery alone are charged and both the first battery and the second battery are connected are estimated, and when the first battery or the second battery becomes discharged, the switch SW is switched so that the charging currents of the first battery and the second battery are connected to each other, otherwise, when the first battery alone becomes charged with a constant voltage, both the first battery and the second battery are connected to each other, otherwise, the ratio of the on time of the switch SW of the first battery alone, the on time of the switch SW of the second battery alone, and the time of the switch SW of both the batteries is controlled so that constant current charging is achieved.

11. The battery system according to claim 10,

the ratio of the on time of the switch SW for the first battery alone, the on time of the switch SW for the second battery alone, and the switch SW time for connecting both batteries is set to a predetermined value.

12. The battery system according to claim 10,

as the ratios of the on time of the switch SW for the first battery alone, the on time of the switch SW for the second battery alone, and the time of the switch SW for connecting both batteries, the ratio of the switch SW for connecting both batteries is set to 0, the time ratio of the switch SW for the first battery is initially set to 1, and the ratio of the switch SW for the first battery is set to (voltage at constant voltage charging of the ac generator-open voltage of the first battery)/(current at constant current charging of the ac generator × polarization resistance) after the first battery has reached the constant current charging completion state.

13. The battery system according to claim 10,

the ratio of the switch SW time for connecting both batteries is set to 0 as the ratio of the switch SW on time for the first battery alone, the switch SW on time for the second battery alone, and the ratio of the switch SW time for connecting both batteries is controlled so that the current at the time of constant current charging of the ac power generator is equal to the polarization voltage of the second battery/(polarization resistance of the second battery (1+ polarization capacitance of the second battery/capacitance of the second battery (F equivalent))) + the polarization voltage of the first battery/polarization resistance of the first battery.

14. The battery system according to claim 1 or 2,

the battery pack has a unit for measuring a voltage and a current of the first battery, a voltage and a current of the second battery, a voltage of the AC generator and a voltage of the auxiliary machine, and a unit for measuring a DC resistance of the first battery, a resistance of the switch SW, a polarization capacitance, a polarization resistance, a polarization voltage, an open-circuit voltage, and a capacitance of the second battery.

15. The battery system of claim 14,

the DC resistance of the battery and the resistance of the switch SW are obtained from the change of the current and the voltage before and after the switch SW is changed from on to off or the switch SW is changed from off to on, and the open-circuit voltage of the battery is obtained as the voltage-DC resistance x current of the battery.

16. The battery system of claim 14,

it is assumed that the polarization voltage is one polarization voltage x (1-measurement time step/(polarization resistance × polarization capacitance)) + measurement current × measurement time step/polarization capacitance before the measurement time.

17. The battery system of claim 15,

the open circuit voltage of the battery which became the steady state was set as the measured open circuit voltage-polarization voltage.

18. The battery system of claim 14,

the relation between the open-circuit voltage of the battery and the state of charge of the battery, which is in a steady state of the battery, is held in advance, the initial state of charge of the battery is obtained from the voltage at the time of system start, the state of charge is updated by adding the value of current integration/the capacity (Ah) of the battery, and the open-circuit voltage in the steady state is obtained from the state of charge.

19. The battery system of claim 14,

the polarization voltage of the battery was set to the measured voltage of the battery-direct current resistance x measured current-open circuit voltage of the battery in a steady state.

20. The battery system according to claim 1 or 2,

the current chargeable in the battery system is transmitted to the upper controller via the communication line.

21. The battery system according to claim 1 or 2,

a charge start signal, a current during constant current charging, and a voltage during constant voltage charging are transmitted from a higher-level controller to the battery system via a communication line.

22. The battery system according to claim 1 or 2,

the time from the start to the end of charging is transmitted from the higher-level controller to the battery system via the communication line together with the start of charging.